Not Applicable
Not Applicable
Not Applicable
1. Field of the Invention
The present invention pertains generally to battery chargers and more particularly to a method for charging batteries which utilizes a variable voltage lid which is responsive to the estimated charge acceptance level to thereby prevent overcharging of the battery as it nears a fully charged state.
2. Description of the Background Art
The process by which a battery is charged determines the relative usable capacity of that battery and to a large degree the service life that can be expected from the battery. Insufficient charging of a battery results in a requisite reduction in battery capacity, wherein the available ampere-hours are inadequate in consideration of the weight, size, and cost of the battery. In contrast, overcharging a battery leads to a reduction in service life for the cells of the battery. Determining a proper charge rate for any battery is complicated by the fact that a fully depleted battery can accept a higher charge rate than a battery which is approaching a state of full-charge, therefore, batteries are typically charged at a variable rate. Unfortunately, the situation is further complicated by the fact that as the battery approaches a fully charged state the charge acceptance drops and charge voltage rises to create an overcharge potential which produces damaging effects on the battery.
Numerous charging methods have been developed, therefore, to provide a charge rate which can fully charge the battery while introducing a limited amount of overcharging. For example, constant current chargers typically generate a constant charge current held within a limited voltage, such that the current drops off as the battery approaches the upper voltage limit of the charger output. FIG. 1 depicts a battery under charge 10, wherein a battery 12 is connected to a voltage source 14 with upper limit VMAX which drives a constant current through the constant current regulator 16 to provide charge current. The charger illustrated in FIG. 1 is a typical example of a CI/CV charger employed in a variety of applications where it supplies a constant current limited by a constant upper voltage limit.
Vehicles often employ CI/CV charging systems which are typically designed to maximize service life by maintaining the battery state-of-charge (SOC) at a moderate level, so as to reduce the deleterious overcharging effects. In battery charging literature and practice, a number of algorithms concerning battery charging provide compromises between service life and performance. A common approach is to maintain the battery at a nominal level of about an 80% state-of-charge (SOC) at all times such that the vehicle power system operates within a narrow SOC range from about 70% to 90%. However, in view of the demands for increases in energy density it is prudent to attempt to maximize the SOC operating window and utilization of the battery.
Numerous misconceptions exist with regard to battery charging which have been promulgated within typical battery charging systems. An application engineer may posit the question xe2x80x9cat what voltage should a specific battery be charged?xe2x80x9d The question is understandable in relation to FIG. 1, however, it is misleading, as are many similar questions and does not lead toward establishing mechanisms for proper charging. To advance the art of charger design toward maximum battery utilization requires re-examination of the underlying charging concepts. Considered in a strict sense, a battery may not be charged by a xe2x80x9cconstant-voltagexe2x80x9d source as it is the concomitant charging current associated with the driving force of the voltage that forces energy storage to occur within the battery. The xe2x80x9cconstant-voltagexe2x80x9d is more correctly the upper limit of the charging voltage which is not exceeded during charging. It will be appreciated that charging at a xe2x80x9cconstant-voltagexe2x80x9d would force unrealistic charge current levels into a depleted battery.
It is beneficial to understand the factors relating to a battery being charged. While undergoing charging, the voltage seen at the terminals of the battery is substantially the sum of three components represented as:
Measured Voltage=Equilibrium Voltage+Polarization Voltage+OhmicVoltageDropxe2x80x83xe2x80x83(1)
wherein the equilibrium voltage is commonly referred to as the open-circuit battery voltage, VOC; the polarization voltage describes the combined effects of concentration and ion/charge-transfer; while the ohmic voltage drop is the voltage drop associated with the ohmic resistance at the given charge current. In contrast to typical electrical components, a battery is an energy storage device that absorbs and provides electrical energy according to an internal electrochemical balance which has an associated reaction voltage that is a dynamic reflection of the xe2x80x9cdriving forcexe2x80x9d function and depends strongly on the past operating history, or time derivatives, experienced by the battery.
FIG. 2 depicts basic charging effects, wherein the battery voltage profile is shown as a function of state-of-charge (SOC) for a series of charging currents 20b through 20f in reference to an equilibrium voltage 20a. The equilibrium voltage 20a is the voltage which would be measured across the open-circuit battery at that point in the charge cycle as represented by the voltage curve if the applied charging current were interrupted or disconnected and equilibrium established. Battery charging current is often expressed as a ratio, C-rate, which expresses the ratio of charging current to nominal battery capacity, I/QN, so that the charge rate may be expressed independently of battery capacity. Charging current curves 20b through 20f identify increasing levels of charging current applied to the battery with 20b at a 0.05C-rate, 20c at a 0.10C-rate, 20d at a 0.33C-rate, 20e at a 0.67C-rate, and 20f at a 1C-rate. It can be seen that during charging, the induced battery voltage exceeds the equilibrium voltage 20a as one would expect in order to force energy into the battery. The curves also indicate that as the battery nears full charge (100% SOC), the battery voltage increases more readily than the equilibrium voltage so as to cause the voltage curves to diverge. In literature, the divergence characteristic of the charge curve from the equilibrium voltage is commonly interpreted as an increase in battery internal resistance as a function of SOC, and simple equivalent circuits and mathematical models are derived accordingly. However, the rationale of such internal resistance concepts are contradictory to the actual chemical and electrochemical nature of a battery. As active materials are converted from lead sulfate, PbSO4 (insulator) in the discharged state in both electrodes to lead dioxide, PbO2 (1.2xc3x9710xe2x88x926 to 2xc3x9710xe2x88x925 xcexa9/m) within the positive electrode and metallic lead Pb (10xe2x88x927 xcexa9/m) within the negative electrode, the overall cell resistance decreases rather than increases. The attendant increase in sulfuric acid concentration that accompanies charging generally causes a minimal increase (less than 10%) on the conductivity of the electrolyte. The electrolyte concentration is typically in the range from 1.250 to 1.280 kg/L. Furthermore, changes to the resistance of metallic parts, e.g., terminals, cell interconnects, lugs, during a single charge cycle is negligible such that ohmic resistance is largely unchanged. Finally it should be appreciated that temperature increases caused by ohmic and joule heating result in further decreases in ohmic resistance within the battery.
It will be appreciated, therefore, that the concept of increasing internal resistance during battery charging is misleading, since resistance levels within the battery do not significantly increase as the state-of-charge increases. In reality, the decrease of charge acceptance is primarily caused by physical blockage of mass transfer as a result of gas entrapment. Mathematically, it is the apparent resistance (xcex94V/I) that is actually increased as a battery is being recharged, but the increase is not due to an increase in electrical resistance.
A constant voltage drop caused by the physical blockage accompanies any specific level of accepted charging current and is referred to as a xe2x80x9cpolarization voltagexe2x80x9d effect. The combined polarization voltage can be expressed by a simplified Tafel correlation that summarizes the polarization and ohmic effects:
xcex7=xcex1+xcex2log(xcfx86)+xcexaxcfx86xe2x80x83xe2x80x83(2)
wherein xcex7 is the combined polarization voltage, xcex1 and xcex2 are the Tafel coefficients, xcexa is the characteristic resistance, and xcfx86 is the equivalent charging current in reference to battery capacity. This correlation is typically valid for instances of charging at low-to-moderate SOC levels which exhibit no overcharging effects, while being charged at a constant-current below a 4C-rate of charge and is typical for the majority of lead-acid batteries. As charging current approaches zero, the mathematical correlation becomes invalid due to the presence of the logarithmic term. In practice, this phenomenon may be described as a minimum amount (or yield) of input energy which is required to polarize or charge the surface layer in order to initiate the process of material conversion.
As a battery approaches a full-charge state (100% SOC), the voltage profile shows a typical characteristic of sharp voltage increase. At this point, the majority of active materials has been converted. The excess charging current is diverted to effect overcharge reactions resulting in the formation of hydrogen and oxygen, known as xe2x80x9cgassingxe2x80x9d or xe2x80x9coutgassingxe2x80x9d, by electrochemically dissociating the water molecules. An increase in the measured battery voltage is induced by the water dissociation process. FIG. 3 is a representation of a typical charge voltage profile showing a single constant-current charge cycle curve 22 along with constituent voltage contributions. The battery equilibrium voltage 24 follows the smooth curve approaching a flat, or nearly flat, region of the curve at 100% SOC. The charging of the battery induces a chemical polarization voltage 26, while the flow of current through the fixed battery resistance results in an ohmic voltage drop 28 which also contributes to the overall charge voltage. The measured voltage generally follows a Tafel relationship up to the gassing point which typically occurs between 50% and 80% SOC for the aforesaid charge rates. It should be recognized that the voltage drop due to ohmic resistance remains relatively constant in view of any specific charge current level. A sudden increase in the measured voltage is exhibited within the upper voltage curve 22 that results in an xe2x80x9covercharge potentialxe2x80x9d 30 being applied to the battery which typically occurs in conjunction with a corresponding decrease in the charge acceptance due to the depletion of active reaction sites and the physical blockage of acid diffusion by the production of gas bubbles. As the overcharge potential increases, the effective charging current that induces chemical and ionic polarization at the surface of electrodes decreases. The battery charging energy becoming predominately consumed by the overcharge reactions rather than electrochemical conversion.
FIG. 4 represents voltages exhibited for various charge currents as a function of state-of-charge (SOC) for a typical battery being charged by means of a constant-current/constant-voltage (CI/CV) charging method. Curves xe2x80x9cAxe2x80x9d, xe2x80x9cBxe2x80x9d, and xe2x80x9cCxe2x80x9d, respectively, are a high, medium, and low rate of charge. The dashed lines 32, 34, 36 starting on the right of curves A, B, C, are representative of fixed voltage levels to which a CI/CV charger is being limited. FIG. 5 illustrates an example of CI/CV charging wherein curve xe2x80x9cAxe2x80x9d from FIG. 4 is limited by the median voltage level 34. The constituent voltages associated with the high current charge curve xe2x80x9cAxe2x80x9d from FIG. 4, wherein the charging voltage of curve xe2x80x9cAxe2x80x9d rises up and is limited by the medium voltage limit 34. The constituent ohmic drop 38 and polarization voltage drop 40 are shown cumulative to the equilibrium voltage 42 of the battery. It will be recognized that although the charging current is being limited by the medium voltage limit 34, there still exists a considerable level of overcharge potential 44 which fosters deleterious outgassing effects. As the charging current decreases as a result of current being limited by the charge voltage limit 34, both the ohmic and polarization voltages decrease accordingly, such that an increasing overcharge current needs to be delivered to the battery in order to maintain a constant voltage. If the supplied initial charge current level is increased, this would precipitate early voltage regulation at lower values for SOC that would result in a slow CV charging step that has an excessively long-tailed charge time profile. Early SOC regulation is the typical method applied to the charging of lithium-based batteries. The charging voltage is set to the equilibrium voltage of a fully charged battery to avoid any amount of overcharging, however as a result, the battery then lacks a sufficient voltage potential to charge rapidly or to reach a 100% state-of-charge.
To reduce the required charging period, a number of designs have employed algorithms which determine a so-called characteristic voltage limit which are applied to produce constant-voltage charging current when the charging voltage reaches that upper limit of voltage. FIG. 6 is illustrative of this approach wherein the median current charging profile xe2x80x9cBxe2x80x9d as depicted in FIG. 4, is limited by the same median voltage threshold 34 so as to reduce the overcharge potential 46 shown in FIG. 6. In following the described teachings of these designs it is apparent that rapid charging may be achieved only at the cost of sacrificing battery longevity due to the high levels of attendant overcharge potential. The setting for the voltage limit characteristically results in a tradeoff between overcharge and charge time. Despite the aforesaid drawbacks of the fixed voltage limited designs, these designs are embodied within the common charging methods applied to the charging of most batteries, in particular lead-acid and nickel-based batteries.
FIG. 7 represents the low current charging curve xe2x80x9cCxe2x80x9d wherein the median voltage charge threshold 34 is never reached and the battery is exposed to the full overcharge potential 48. This method of regulation has been implemented within charger designs, such as those which employ negative delta-V measurements within sealed lead-acid and metal hydride batteries. The voltage limit within these designs provides for a means of detecting erroneous high-current charging or runaway conditions, but does not limit the overcharging which occurs from normal operation.
As shown above, overcharging a battery is an inevitable result of employing a constant-voltage charge step. These simple illustrations have additionally shown that (1) it is meaningless to specify a xe2x80x9ccharging voltagexe2x80x9d because CV charging and/or overcharge can occur at any SOC depending only on the applied current; and (2) a high charging voltage does not (and will not) improve the effectiveness of battery charging but only provides a more complete charge return by means of excessive overcharge.
It will be appreciated that in each instance of voltage limitations imposed on the charging current there was an appreciable level of undesirable battery overcharging. Contemplating the use of the high limit 32 or the low limit 36 as shown in FIG. 4, it will be understood that each of these limits results in either overcharging the battery or restricting the battery from reaching, or quickly reaching, a full state of charge. It will be appreciated, therefore, that specifying a xe2x80x9ccharging voltagexe2x80x9d, as illustrated above, does not mitigate the problems of slow charge rates or the attendant battery damage caused by overcharge potentials, while increasing the voltage threshold can speed charging only by inducing increased overcharge potentials.
FIG. 8 illustrates the hypothetical case of charging a battery in a true constant-voltage charging process. During the initial stages of charging the fixed charge voltage 50 of the charger attempts to drive an unlimited amount of electrical current or electrons into the battery forcing a rapid movement of ionic species toward the electrode-electrolyte interface. Comprising the voltage drop between the equilibrium voltage 52 and the charge voltage 50, are the ohmic voltage drop 54, the polarization voltage drop 56, and the overcharging potential 58. As the material conversion near the electrolyte interface approaches saturation, the battery charge acceptance decreases drastically due to the lack of chemical (or ionic) and mass-transfer driving force. Thus, the charge current drops rapidly to a low level (yet sufficiently high to maintain a constant voltage) wherein a great portion of the input energy is wasted, and serves only to promote the overcharge reactions.
It should also be noted that typical vehicular charging systems operate in a similar fashion as a CI/CV charging system except that the charging current in the CI phase is limited by the maximum power output of the generator. Instead of a CI phase, a rapidly decreasing input current profile is observed as the battery voltage increases, i.e., Vxc3x97I=constant, which is often mistaken as CV charging only.
A method of stepped-current charging is often employed for charging batteries which promotes a relaxation of built-up overcharge gasses to provide increased charge efficiency. A stepped-current charger is exemplified in U.S. Pat. No. 5,561,360 issued Oct. 1, 1996 to Ayres et al. which is incorporated herein by reference. FIG. 9 shows a block diagram 60 of a typical charger which employs a controller. Battery 62 is charged by programmable current source 64 in a stepped current mode controlled by controller 66 employing a voltage limit set for VMAX. The voltage limit scales upwardly as the charge acceptance of the battery declines so as to increase the driving force (overvoltage) to aid energy input. Graphs for this type of charging are shown in FIG. 10 and FIG. 11. The voltage, current and gas flow produced by this method during charging are shown in FIG. 10 with the associated applied voltage as a function of state-of-charge being shown in FIG. 11. It will be appreciated that the method produces high levels of overcharging within the battery. The maximum charging current typical of this type of charging is about C/3 (C=battery capacity in ampere hours) which results in reasonable performance with overcharging of about 105-125%, wherein overcharging is specified as a ratio of input ampere-hour capacity to total discharge ampere-hour capacity (100% depth-of-discharge of the previous discharge cycle). The overcharge ratio is often evaluated within the industry using the nominal capacity of the test battery as the denominator. For example, for a 50-Ah battery, a fully discharged battery may be recharged by returning about 62.5 Ah (overcharge ratio of 125%) to the battery, wherein the ampere-hour capacity being returned is often used as a termination setting for ampere-hour integration. As illustrated in FIG. 10, overcharging is significantly reduced during the initial high-current steps, yet with continuous and progressively increasing overcharge being needed to force the battery voltage to reach the increasing voltage levels and effect the subsequent current regulation steps. As the charging current reduces to very low levels (e.g., as indicated by the last two current steps in FIG. 11), the battery voltage is unable to reach the required voltage limit which requires that the charging procedure additionally impose a time-based or ampere-hour based limitation such as illustrated in FIG. 10. It will be appreciated that the charger provides beneficial reductions of outgassing at low charge rates, however, the moderate levels of outgassing which still remain and the additional limits of time and/or ampere-hour which must be imposed are significant detractors to applicability of the approach.
Therefore a need exists for a method of charging a battery that does not require striking a win-lose tradeoff between overcharging and charging time. The methods described for the present invention satisfy those needs, as well as others, while overcoming deficiencies inherent within previous charging methods.
The present invention provides a battery charging method which is suitable for charging any battery system rapidly from an arbitrarily low state-of-charge while minimizing overcharging as the SOC progresses toward full charge. A major adverse effect of overcharging and side reactions is a deterioration of the service life for the battery as the result of changes which are exemplified by water loss and the shedding of active material. The charging method comprises a technique based on the macroscopic relationship of electrical and electrochemical behavior of a battery under charge that is suitable for use with charging a variety of batteries, most notably lead acid batteries, and may be employed within a variety of charging systems, such as chargers similar to that depicted in FIG. 9.
The charging method determines an optimum variable voltage charging lid locus which is based upon the charge-acceptance capability of the battery that generally follows a decreasing curve which may be associated to the actual state-of-charge (SOC) within the battery at any point in time. The optimum variable voltage charging lid may alternatively be approximated by either a similarly shaped relationship, or as a series of stepped-down voltage limits which approximate the shape of the variable voltage charging lid curve. The degree of matching required being determined by the level of overcharging to be allowed within a particular application. The variable voltage lid commences when the battery nears a full-state of charge and it provides moderation of the applied current which facilitates rapid battery charging without the high overcharge potentials typified by CI and CI/CV battery charging approaches. The battery charging method of the present invention is suitable for charging batteries of various chemistries, such as lead-acid, nickel-based (i.e. nickel-cadmium, nickel-metal-hydride) and lithium-based batteries.
An object of the invention is to provide a method of charging batteries that does not subject the batteries to high overcharge potentials.
Another object of the invention is to provide a charge method that can be utilized for rapidly charging batteries without attendant damage thereto.
Another object of the invention is to provide a charging method that can be implemented easily and at low-cost within a variety of charging systems.
Another object of the invention is to provide a charging method that operates efficiently within a highly dynamic state-of-charge environment wherein the battery vacillates between discharging and recharging.
Further objects and advantages of the invention will be brought out in the following portions of the specification, wherein the detailed description is for the purpose of fully disclosing preferred embodiments of the invention without placing limitations thereon.